This retrospective study evaluated the progression of deformity after posterior fusion by reviewing 63 consecutive patients with idiopathic scoliosis who were all in Risser sign 0 at the time of surgery. All patients were observed beyond the time of skeletal maturity. Average follow-up time was 9 years and 8 months (range, 5-16 years). To investigate the risk factors for the crankshaft phenomenon after posterior fusion and to build a model for predicting the probability of curve progression until maturation of growth. There remains considerable controversy concerning the incidence, risk factors, and necessity of combined anterior fusion to prevent the crankshaft phenomenon in patients who are skeletally immature. Serial radiographs were measured for Cobb angle, apical rotation according to Perdriolle, and apical rib-vertebra angle of Mehta. Multivariate and univariate logistic regression analysis was performed using seven potential predictors as independent variables and Cobb angle progression and rotational progression as dependent variables. Average progression of deformity was 3 degrees Cobb angle (range, -8-16 degrees) and 3 degrees Perdriolle rotation (range, -9-17 degrees). Progression of deformity more than 5 degrees of either Cobb angle or rotation was observed in 22 (35%) of 63 curves with 7 (11%) of 63 curves greater than 10 degrees. Chronologic age and skeletal age were found to be significantly associated with progression of deformity in univariate analysis. In multivariate analysis, only skeletal age seemed to be independently prognostic. The authors tried to build the logistic model using the three factors of chronologic age, skeletal age, and apical rib-vertebra angle. This model correctly classified 81% of all patients as progressive or nonprogressive. The positive predictive value was 90%. The results showed that patients with chronologic age of 11 years of younger, especially those with a skeletal age of 10 years or younger, had a high estimated probability of progression of deformity. The progression was fairly moderate, however, with an average Cobb angle of 9 degrees and average rotation of 7 degrees, which neither the patients nor the surgeon believed was of such magnitude as to warrant routine combined anterior fusion.